| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.23143277929728592, | |
| "eval_steps": 10, | |
| "global_step": 550, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004207868714496108, | |
| "grad_norm": 3.2594902515411377, | |
| "learning_rate": 0.00039272727272727273, | |
| "loss": 0.6971, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004207868714496108, | |
| "eval_accuracy": 0.5914159417152405, | |
| "eval_loss": 0.7148427963256836, | |
| "eval_runtime": 585.2226, | |
| "eval_samples_per_second": 8.122, | |
| "eval_steps_per_second": 2.032, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008415737428992216, | |
| "grad_norm": 22.503503799438477, | |
| "learning_rate": 0.0003854545454545455, | |
| "loss": 0.7436, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008415737428992216, | |
| "eval_accuracy": 0.5914159417152405, | |
| "eval_loss": 1.42451012134552, | |
| "eval_runtime": 577.2378, | |
| "eval_samples_per_second": 8.234, | |
| "eval_steps_per_second": 2.06, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.012623606143488323, | |
| "grad_norm": 1.2163587808609009, | |
| "learning_rate": 0.0003781818181818182, | |
| "loss": 0.9446, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.012623606143488323, | |
| "eval_accuracy": 0.4085840582847595, | |
| "eval_loss": 0.7073472738265991, | |
| "eval_runtime": 581.0519, | |
| "eval_samples_per_second": 8.18, | |
| "eval_steps_per_second": 2.046, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.016831474857984433, | |
| "grad_norm": 2.1413767337799072, | |
| "learning_rate": 0.0003709090909090909, | |
| "loss": 0.6945, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.016831474857984433, | |
| "eval_accuracy": 0.5914159417152405, | |
| "eval_loss": 0.7796906232833862, | |
| "eval_runtime": 591.5892, | |
| "eval_samples_per_second": 8.034, | |
| "eval_steps_per_second": 2.01, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.021039343572480537, | |
| "grad_norm": 0.8763795495033264, | |
| "learning_rate": 0.00036363636363636367, | |
| "loss": 0.653, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.021039343572480537, | |
| "eval_accuracy": 0.5867872834205627, | |
| "eval_loss": 0.6816809773445129, | |
| "eval_runtime": 588.5616, | |
| "eval_samples_per_second": 8.076, | |
| "eval_steps_per_second": 2.02, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.025247212286976645, | |
| "grad_norm": 0.9567063450813293, | |
| "learning_rate": 0.0003563636363636364, | |
| "loss": 0.6479, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.025247212286976645, | |
| "eval_accuracy": 0.8102251291275024, | |
| "eval_loss": 0.5443565845489502, | |
| "eval_runtime": 581.8352, | |
| "eval_samples_per_second": 8.169, | |
| "eval_steps_per_second": 2.044, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.029455081001472753, | |
| "grad_norm": 4.155121803283691, | |
| "learning_rate": 0.0003490909090909091, | |
| "loss": 0.502, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.029455081001472753, | |
| "eval_accuracy": 0.5952030420303345, | |
| "eval_loss": 1.1268996000289917, | |
| "eval_runtime": 583.4269, | |
| "eval_samples_per_second": 8.147, | |
| "eval_steps_per_second": 2.038, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.033662949715968865, | |
| "grad_norm": 3.364042282104492, | |
| "learning_rate": 0.0003418181818181818, | |
| "loss": 0.6052, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.033662949715968865, | |
| "eval_accuracy": 0.8194824457168579, | |
| "eval_loss": 0.6350404024124146, | |
| "eval_runtime": 585.0166, | |
| "eval_samples_per_second": 8.125, | |
| "eval_steps_per_second": 2.032, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.03787081843046497, | |
| "grad_norm": 1.5822795629501343, | |
| "learning_rate": 0.00033454545454545456, | |
| "loss": 0.3917, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.03787081843046497, | |
| "eval_accuracy": 0.5914159417152405, | |
| "eval_loss": 1.0552942752838135, | |
| "eval_runtime": 579.8411, | |
| "eval_samples_per_second": 8.197, | |
| "eval_steps_per_second": 2.051, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.042078687144961074, | |
| "grad_norm": 1.6902482509613037, | |
| "learning_rate": 0.0003272727272727273, | |
| "loss": 0.648, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.042078687144961074, | |
| "eval_accuracy": 0.8116979002952576, | |
| "eval_loss": 0.7334651947021484, | |
| "eval_runtime": 577.8009, | |
| "eval_samples_per_second": 8.226, | |
| "eval_steps_per_second": 2.058, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.046286555859457186, | |
| "grad_norm": 1.05351984500885, | |
| "learning_rate": 0.00032, | |
| "loss": 0.431, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.046286555859457186, | |
| "eval_accuracy": 0.8480959534645081, | |
| "eval_loss": 0.39465391635894775, | |
| "eval_runtime": 576.7671, | |
| "eval_samples_per_second": 8.241, | |
| "eval_steps_per_second": 2.061, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.05049442457395329, | |
| "grad_norm": 0.6553493738174438, | |
| "learning_rate": 0.00031272727272727273, | |
| "loss": 0.5428, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.05049442457395329, | |
| "eval_accuracy": 0.8729223608970642, | |
| "eval_loss": 0.38810551166534424, | |
| "eval_runtime": 579.7258, | |
| "eval_samples_per_second": 8.199, | |
| "eval_steps_per_second": 2.051, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.0547022932884494, | |
| "grad_norm": 0.42037296295166016, | |
| "learning_rate": 0.0003054545454545455, | |
| "loss": 0.34, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.0547022932884494, | |
| "eval_accuracy": 0.8840731978416443, | |
| "eval_loss": 0.35628741979599, | |
| "eval_runtime": 582.4004, | |
| "eval_samples_per_second": 8.161, | |
| "eval_steps_per_second": 2.042, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.05891016200294551, | |
| "grad_norm": 0.47820377349853516, | |
| "learning_rate": 0.0002981818181818182, | |
| "loss": 0.398, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.05891016200294551, | |
| "eval_accuracy": 0.8842836022377014, | |
| "eval_loss": 0.3024275004863739, | |
| "eval_runtime": 581.4736, | |
| "eval_samples_per_second": 8.174, | |
| "eval_steps_per_second": 2.045, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.06311803071744161, | |
| "grad_norm": 0.9904082417488098, | |
| "learning_rate": 0.0002909090909090909, | |
| "loss": 0.1834, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06311803071744161, | |
| "eval_accuracy": 0.87250155210495, | |
| "eval_loss": 0.38005706667900085, | |
| "eval_runtime": 580.8975, | |
| "eval_samples_per_second": 8.182, | |
| "eval_steps_per_second": 2.047, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06732589943193773, | |
| "grad_norm": 1.6720079183578491, | |
| "learning_rate": 0.0002836363636363637, | |
| "loss": 0.5052, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.06732589943193773, | |
| "eval_accuracy": 0.9114243388175964, | |
| "eval_loss": 0.23850664496421814, | |
| "eval_runtime": 581.0949, | |
| "eval_samples_per_second": 8.179, | |
| "eval_steps_per_second": 2.046, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.07153376814643383, | |
| "grad_norm": 1.5653446912765503, | |
| "learning_rate": 0.0002763636363636364, | |
| "loss": 0.3953, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.07153376814643383, | |
| "eval_accuracy": 0.8544077277183533, | |
| "eval_loss": 0.41211748123168945, | |
| "eval_runtime": 581.5472, | |
| "eval_samples_per_second": 8.173, | |
| "eval_steps_per_second": 2.045, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.07574163686092994, | |
| "grad_norm": 1.7199907302856445, | |
| "learning_rate": 0.0002690909090909091, | |
| "loss": 0.407, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.07574163686092994, | |
| "eval_accuracy": 0.9044813513755798, | |
| "eval_loss": 0.26600247621536255, | |
| "eval_runtime": 580.7861, | |
| "eval_samples_per_second": 8.184, | |
| "eval_steps_per_second": 2.047, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.07994950557542604, | |
| "grad_norm": 2.303934335708618, | |
| "learning_rate": 0.00026181818181818185, | |
| "loss": 0.3571, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.07994950557542604, | |
| "eval_accuracy": 0.8844940066337585, | |
| "eval_loss": 0.3394128680229187, | |
| "eval_runtime": 584.97, | |
| "eval_samples_per_second": 8.125, | |
| "eval_steps_per_second": 2.033, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.08415737428992215, | |
| "grad_norm": 2.5085370540618896, | |
| "learning_rate": 0.00025454545454545456, | |
| "loss": 0.2747, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08415737428992215, | |
| "eval_accuracy": 0.9147906303405762, | |
| "eval_loss": 0.2246033400297165, | |
| "eval_runtime": 585.8187, | |
| "eval_samples_per_second": 8.113, | |
| "eval_steps_per_second": 2.03, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08836524300441827, | |
| "grad_norm": 4.109118461608887, | |
| "learning_rate": 0.00024727272727272727, | |
| "loss": 0.2863, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.08836524300441827, | |
| "eval_accuracy": 0.9244687557220459, | |
| "eval_loss": 0.2438182830810547, | |
| "eval_runtime": 582.1202, | |
| "eval_samples_per_second": 8.165, | |
| "eval_steps_per_second": 2.043, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.09257311171891437, | |
| "grad_norm": 0.5757103562355042, | |
| "learning_rate": 0.00024, | |
| "loss": 0.2334, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.09257311171891437, | |
| "eval_accuracy": 0.922154426574707, | |
| "eval_loss": 0.21005088090896606, | |
| "eval_runtime": 579.1924, | |
| "eval_samples_per_second": 8.206, | |
| "eval_steps_per_second": 2.053, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.09678098043341048, | |
| "grad_norm": 0.36710911989212036, | |
| "learning_rate": 0.00023272727272727271, | |
| "loss": 0.1744, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.09678098043341048, | |
| "eval_accuracy": 0.8529350161552429, | |
| "eval_loss": 0.41234469413757324, | |
| "eval_runtime": 589.1406, | |
| "eval_samples_per_second": 8.068, | |
| "eval_steps_per_second": 2.018, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.10098884914790658, | |
| "grad_norm": 3.2360424995422363, | |
| "learning_rate": 0.00022545454545454545, | |
| "loss": 0.1948, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.10098884914790658, | |
| "eval_accuracy": 0.9253103137016296, | |
| "eval_loss": 0.22991585731506348, | |
| "eval_runtime": 587.1353, | |
| "eval_samples_per_second": 8.095, | |
| "eval_steps_per_second": 2.025, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.1051967178624027, | |
| "grad_norm": 0.21486328542232513, | |
| "learning_rate": 0.00021818181818181818, | |
| "loss": 0.2382, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1051967178624027, | |
| "eval_accuracy": 0.9322533011436462, | |
| "eval_loss": 0.27035772800445557, | |
| "eval_runtime": 587.222, | |
| "eval_samples_per_second": 8.094, | |
| "eval_steps_per_second": 2.025, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1094045865768988, | |
| "grad_norm": 1.8511269092559814, | |
| "learning_rate": 0.0002109090909090909, | |
| "loss": 0.219, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.1094045865768988, | |
| "eval_accuracy": 0.9137386679649353, | |
| "eval_loss": 0.3539877235889435, | |
| "eval_runtime": 585.6282, | |
| "eval_samples_per_second": 8.116, | |
| "eval_steps_per_second": 2.03, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.11361245529139491, | |
| "grad_norm": 0.11713656038045883, | |
| "learning_rate": 0.00020363636363636363, | |
| "loss": 0.1122, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.11361245529139491, | |
| "eval_accuracy": 0.9339364767074585, | |
| "eval_loss": 0.2783205509185791, | |
| "eval_runtime": 585.5155, | |
| "eval_samples_per_second": 8.118, | |
| "eval_steps_per_second": 2.031, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.11782032400589101, | |
| "grad_norm": 2.4492409229278564, | |
| "learning_rate": 0.00019636363636363636, | |
| "loss": 0.1902, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.11782032400589101, | |
| "eval_accuracy": 0.9322533011436462, | |
| "eval_loss": 0.28133705258369446, | |
| "eval_runtime": 587.357, | |
| "eval_samples_per_second": 8.092, | |
| "eval_steps_per_second": 2.024, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.12202819272038712, | |
| "grad_norm": 0.09383056312799454, | |
| "learning_rate": 0.0001890909090909091, | |
| "loss": 0.1279, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.12202819272038712, | |
| "eval_accuracy": 0.9297285676002502, | |
| "eval_loss": 0.22569426894187927, | |
| "eval_runtime": 586.2579, | |
| "eval_samples_per_second": 8.107, | |
| "eval_steps_per_second": 2.028, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.12623606143488322, | |
| "grad_norm": 1.5099377632141113, | |
| "learning_rate": 0.00018181818181818183, | |
| "loss": 0.168, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.12623606143488322, | |
| "eval_accuracy": 0.9347780346870422, | |
| "eval_loss": 0.2831152081489563, | |
| "eval_runtime": 587.962, | |
| "eval_samples_per_second": 8.084, | |
| "eval_steps_per_second": 2.022, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13044393014937933, | |
| "grad_norm": 16.816967010498047, | |
| "learning_rate": 0.00017454545454545454, | |
| "loss": 0.1351, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.13044393014937933, | |
| "eval_accuracy": 0.9293078184127808, | |
| "eval_loss": 0.3322593569755554, | |
| "eval_runtime": 586.8235, | |
| "eval_samples_per_second": 8.1, | |
| "eval_steps_per_second": 2.026, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.13465179886387546, | |
| "grad_norm": 0.6043083667755127, | |
| "learning_rate": 0.00016727272727272728, | |
| "loss": 0.0422, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.13465179886387546, | |
| "eval_accuracy": 0.9427729845046997, | |
| "eval_loss": 0.28099876642227173, | |
| "eval_runtime": 587.4493, | |
| "eval_samples_per_second": 8.091, | |
| "eval_steps_per_second": 2.024, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.13885966757837157, | |
| "grad_norm": 16.680456161499023, | |
| "learning_rate": 0.00016, | |
| "loss": 0.1904, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.13885966757837157, | |
| "eval_accuracy": 0.9349884390830994, | |
| "eval_loss": 0.3474605977535248, | |
| "eval_runtime": 579.304, | |
| "eval_samples_per_second": 8.205, | |
| "eval_steps_per_second": 2.052, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.14306753629286767, | |
| "grad_norm": 0.6815859079360962, | |
| "learning_rate": 0.00015272727272727275, | |
| "loss": 0.0864, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.14306753629286767, | |
| "eval_accuracy": 0.9438249468803406, | |
| "eval_loss": 0.3012893497943878, | |
| "eval_runtime": 588.316, | |
| "eval_samples_per_second": 8.079, | |
| "eval_steps_per_second": 2.021, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.14727540500736377, | |
| "grad_norm": 0.01422311831265688, | |
| "learning_rate": 0.00014545454545454546, | |
| "loss": 0.0198, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.14727540500736377, | |
| "eval_accuracy": 0.9335156679153442, | |
| "eval_loss": 0.38236290216445923, | |
| "eval_runtime": 589.516, | |
| "eval_samples_per_second": 8.063, | |
| "eval_steps_per_second": 2.017, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.15148327372185988, | |
| "grad_norm": 2.280247211456299, | |
| "learning_rate": 0.0001381818181818182, | |
| "loss": 0.2155, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.15148327372185988, | |
| "eval_accuracy": 0.9463496804237366, | |
| "eval_loss": 0.3106628656387329, | |
| "eval_runtime": 581.9576, | |
| "eval_samples_per_second": 8.167, | |
| "eval_steps_per_second": 2.043, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.15569114243635598, | |
| "grad_norm": 15.617796897888184, | |
| "learning_rate": 0.00013090909090909093, | |
| "loss": 0.2275, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.15569114243635598, | |
| "eval_accuracy": 0.9450873136520386, | |
| "eval_loss": 0.2654193341732025, | |
| "eval_runtime": 582.2813, | |
| "eval_samples_per_second": 8.163, | |
| "eval_steps_per_second": 2.042, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.1598990111508521, | |
| "grad_norm": 9.974563598632812, | |
| "learning_rate": 0.00012363636363636364, | |
| "loss": 0.1118, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.1598990111508521, | |
| "eval_accuracy": 0.9421418309211731, | |
| "eval_loss": 0.2898730933666229, | |
| "eval_runtime": 580.2758, | |
| "eval_samples_per_second": 8.191, | |
| "eval_steps_per_second": 2.049, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.1641068798653482, | |
| "grad_norm": 1.2296732664108276, | |
| "learning_rate": 0.00011636363636363636, | |
| "loss": 0.0258, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.1641068798653482, | |
| "eval_accuracy": 0.9570797681808472, | |
| "eval_loss": 0.18523547053337097, | |
| "eval_runtime": 580.5297, | |
| "eval_samples_per_second": 8.187, | |
| "eval_steps_per_second": 2.048, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.1683147485798443, | |
| "grad_norm": 1.1257351636886597, | |
| "learning_rate": 0.00010909090909090909, | |
| "loss": 0.0816, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.1683147485798443, | |
| "eval_accuracy": 0.9535030722618103, | |
| "eval_loss": 0.18315376341342926, | |
| "eval_runtime": 582.012, | |
| "eval_samples_per_second": 8.166, | |
| "eval_steps_per_second": 2.043, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.17252261729434043, | |
| "grad_norm": 0.1417863965034485, | |
| "learning_rate": 0.00010181818181818181, | |
| "loss": 0.1385, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.17252261729434043, | |
| "eval_accuracy": 0.9606564044952393, | |
| "eval_loss": 0.1722693145275116, | |
| "eval_runtime": 583.7175, | |
| "eval_samples_per_second": 8.143, | |
| "eval_steps_per_second": 2.037, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.17673048600883653, | |
| "grad_norm": 0.05042952299118042, | |
| "learning_rate": 9.454545454545455e-05, | |
| "loss": 0.1194, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.17673048600883653, | |
| "eval_accuracy": 0.9391962885856628, | |
| "eval_loss": 0.26152685284614563, | |
| "eval_runtime": 583.6437, | |
| "eval_samples_per_second": 8.144, | |
| "eval_steps_per_second": 2.037, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.18093835472333264, | |
| "grad_norm": 0.06302843242883682, | |
| "learning_rate": 8.727272727272727e-05, | |
| "loss": 0.2722, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.18093835472333264, | |
| "eval_accuracy": 0.9667578339576721, | |
| "eval_loss": 0.1336488574743271, | |
| "eval_runtime": 585.4585, | |
| "eval_samples_per_second": 8.118, | |
| "eval_steps_per_second": 2.031, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.18514622343782874, | |
| "grad_norm": 1.160020112991333, | |
| "learning_rate": 8e-05, | |
| "loss": 0.1969, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.18514622343782874, | |
| "eval_accuracy": 0.9520303010940552, | |
| "eval_loss": 0.1606164574623108, | |
| "eval_runtime": 582.6331, | |
| "eval_samples_per_second": 8.158, | |
| "eval_steps_per_second": 2.041, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.18935409215232485, | |
| "grad_norm": 0.8594697713851929, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.109, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.18935409215232485, | |
| "eval_accuracy": 0.9610772132873535, | |
| "eval_loss": 0.13081230223178864, | |
| "eval_runtime": 598.3887, | |
| "eval_samples_per_second": 7.943, | |
| "eval_steps_per_second": 1.987, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.19356196086682095, | |
| "grad_norm": 0.1255054622888565, | |
| "learning_rate": 6.545454545454546e-05, | |
| "loss": 0.1662, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.19356196086682095, | |
| "eval_accuracy": 0.9657058715820312, | |
| "eval_loss": 0.1277003139257431, | |
| "eval_runtime": 596.8696, | |
| "eval_samples_per_second": 7.963, | |
| "eval_steps_per_second": 1.992, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.19776982958131706, | |
| "grad_norm": 0.09831862151622772, | |
| "learning_rate": 5.818181818181818e-05, | |
| "loss": 0.0393, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.19776982958131706, | |
| "eval_accuracy": 0.964022696018219, | |
| "eval_loss": 0.12812593579292297, | |
| "eval_runtime": 594.9871, | |
| "eval_samples_per_second": 7.988, | |
| "eval_steps_per_second": 1.998, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.20197769829581316, | |
| "grad_norm": 0.08306915313005447, | |
| "learning_rate": 5.090909090909091e-05, | |
| "loss": 0.1268, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.20197769829581316, | |
| "eval_accuracy": 0.9644435048103333, | |
| "eval_loss": 0.13266168534755707, | |
| "eval_runtime": 586.2759, | |
| "eval_samples_per_second": 8.107, | |
| "eval_steps_per_second": 2.028, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "grad_norm": 0.08147989958524704, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.0548, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "eval_accuracy": 0.9646539092063904, | |
| "eval_loss": 0.14851805567741394, | |
| "eval_runtime": 591.656, | |
| "eval_samples_per_second": 8.033, | |
| "eval_steps_per_second": 2.01, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.2103934357248054, | |
| "grad_norm": 0.6941895484924316, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.0484, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2103934357248054, | |
| "eval_accuracy": 0.9629707336425781, | |
| "eval_loss": 0.16351090371608734, | |
| "eval_runtime": 589.9697, | |
| "eval_samples_per_second": 8.056, | |
| "eval_steps_per_second": 2.015, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2146013044393015, | |
| "grad_norm": 0.04926018416881561, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 0.022, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.2146013044393015, | |
| "eval_accuracy": 0.964022696018219, | |
| "eval_loss": 0.15824884176254272, | |
| "eval_runtime": 597.3025, | |
| "eval_samples_per_second": 7.957, | |
| "eval_steps_per_second": 1.991, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.2188091731537976, | |
| "grad_norm": 0.043105900287628174, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 0.1426, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.2188091731537976, | |
| "eval_accuracy": 0.9642331004142761, | |
| "eval_loss": 0.15840177237987518, | |
| "eval_runtime": 596.1264, | |
| "eval_samples_per_second": 7.973, | |
| "eval_steps_per_second": 1.995, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.22301704186829371, | |
| "grad_norm": 0.04364515841007233, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 0.0611, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.22301704186829371, | |
| "eval_accuracy": 0.9636019468307495, | |
| "eval_loss": 0.1682334691286087, | |
| "eval_runtime": 590.7102, | |
| "eval_samples_per_second": 8.046, | |
| "eval_steps_per_second": 2.013, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.22722491058278982, | |
| "grad_norm": 0.9690969586372375, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 0.0668, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.22722491058278982, | |
| "eval_accuracy": 0.9633915424346924, | |
| "eval_loss": 0.16814225912094116, | |
| "eval_runtime": 593.6369, | |
| "eval_samples_per_second": 8.007, | |
| "eval_steps_per_second": 2.003, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.23143277929728592, | |
| "grad_norm": 0.053498703986406326, | |
| "learning_rate": 0.0, | |
| "loss": 0.0471, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.23143277929728592, | |
| "eval_accuracy": 0.964022696018219, | |
| "eval_loss": 0.16578398644924164, | |
| "eval_runtime": 583.1222, | |
| "eval_samples_per_second": 8.151, | |
| "eval_steps_per_second": 2.039, | |
| "step": 550 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 550, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.208452205931052e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |